In December, five new articles entered our top ten most read list. The Top 10 is compiled based on full text views, including downloads of the full article PDFs.
ChatGPT (GPT-4) versus doctors on complex cases of the Swedish family medicine specialist examination: an observational comparative study
Given increasing interest in artificial intelligence (AI), it is not surprising that Arvidsson et al.’s study using ChatGPT was highly viewed last month. ChatGPT is a well-known online application which can generate text in response to questioning. In this study, the performance of ChatGPT was compared to real doctors in assessing complex medical cases from the Swedish family medicine specialist examination. Responses from three groups were compared: (i) ChatGPT (version GPT-4), (ii) a random real-life exam response, (iii) example of a top-tier response from the Swedish Association of General Practice. The researchers found that GPT-4 scored significantly lower than either doctor group.
Profiles and associated factors of prenatal psychological symptoms and their association with health-related quality of life among pregnant women: a cross-sectional study
Huo et al. aimed to identify and cluster prenatal psychological symptoms, identify associated demographic factors, and compare health-related quality of life across different symptom profiles. Based on mean scores of depression, anxiety and stress, prenatal psychological symptoms were clustered into three profiles: low, moderate, severe. Several factors were significantly associated with moderate to severe symptoms such as younger age and a poorer relationship with one’s partner. The authors also found an inverse association between the severity of prenatal psychological symptoms and health-related quality of life.
Evaluating transportability of overall survival estimates from US to UK populations receiving first-line treatment for advanced non-small cell lung cancer: a retrospective cohort study
This study by Kent et al. explores transportability of evidence which is the concept that data from one population can be extended to another population. In this case, the researchers compared data from the USA and UK regarding patients with advanced non-small cell lung cancer (aNSCLC) initiating first-line treatment. The authors found that patient characteristics and overall survival data were similar between groups within 12 months of initiating first-line treatment, with some differences only appearing after the 1 year mark. The authors concluded that within this population, US data have the potential to be used where UK data are unavailable or sparse.